Search

Abstract: This paper is concerned with distributed sensors to measure the response of beam and plate structures. The design of modal sensors for beam structures is well established. The design for plate structures with constant thickness sensors may be achieved by optimizing the shape of the sensor boundary, or by optimizing the effectiveness of the electrode. Most applications consider vibration control, but this paper is concerned with structural health monitoring, where the sensor output is made sensitive to changes in key stiffness parameters, for example in joints. The procedure is based on finite element models of the structures, and thus distributed transducers may be designed for arbitrary beam and plate structures. Simulated examples of a beam and a plate excited by a rotating machine are used to demonstrate the approach. [Copyright &y& Elsevier]

Abstract: The use of damage-sensitive features to evaluate structural condition or health is a very critical aspect of structural health monitoring. The purpose of this paper is to investigate the potential of two different damage-sensitive features for detecting damage. Different damage scenarios are simulated on a large-scale laboratory structure and a three-span highway bridge for demonstration. The features presented in this paper are the modal flexibility-based deflection and curvature both of which are obtained directly from dynamic properties. In the literature, flexibility associated with mode shapes and mode shapes curvatures have been mostly explored. In this study, multi-input–multi-output dynamic data are used to obtain modal flexibility, which is a close approximation to the actual flexibility. A main novelty is that the curvature is calculated from the deflected shapes using the modal flexibility as opposed to using modal vectors. In this paper, the theory of the methodology is explained and then experimental studies and results are presented. For the experimental studies, the laboratory specimen and the three-span bridge were gradually damaged. It is shown that both deflection and curvature are conceptual and physically meaningful features for damage detection and localization. The issues and the requirements for these features to perform successfully are also presented. [Copyright &y& Elsevier]

• This paper presents a deep learning network that inspects ultrasonic signals for defect detection. • The network is an adaptation of WaveNet, hence is based on causal dilated convolutional neural networks and residual blocks. • The network is shown to outperform both a widely used conventional analysis method and some competing deep learning algorithms. • The validation was performed on two datasets, one generated via finite element simulations on a steel plate, the other by actual experiments on a composite plate. • Significant improvements over the conventional method are obtained especially when testing occurs at temperatures well outside the range in the training set of signals. This paper presents a deep learning network that performs automatic detection of defects by inspecting full ultrasonic guided wave signals excited in plate structures. The findings show that the algorithm, which is an adaptation of WaveNet, and hence is based on causal dilated convolutional neural networks, is effectively able to learn features and/or patterns related to the presence of waves scattered from damage, thus eliminating the need for any feature engineering to be performed by human operators. The network outperformed the widely used conventional approach that combines the optimal baseline selection and baseline signal stretch compensation methods when tested on two different datasets. The first dataset consisted of finite element simulated Lamb wave signals acquired in a pitch-catch configuration on a steel plate across a 50 °C range of temperature variations, and the second was a publicly available experimental dataset of Lamb wave signals also acquired in pitch-catch mode on a composite plate with a 40 °C range of variations. The improvements over the conventional approach are particularly encouraging when analyzing signals at temperatures well outside the temperature range available in the set of baseline signals, hence suggesting that this class of algorithms can complement or substitute existing methods, especially when testing occurs at unseen environmental and operational conditions, or when the effects of sensor drift make conventional methods less effective. [ABSTRACT FROM AUTHOR]

• Population-based SHM allows knowledge transfer between structures of a population. • Collecting manifolds of potential structural states, a fibre bundle is defined. • The normal-condition cross section is of particular interest in the SHM discipline. • A graph representation of structures in combination with graph neural networks is used to construct the section. • Method is demonstrated successfully on a population of simulated structures. One of the requirements of the population-based approach to Structural Health Monitoring (SHM) proposed in the earlier papers in this sequence, is that structures be represented by points in an abstract space. Furthermore, these spaces should be metric spaces in a loose sense; i.e. there should be some measure of distance applicable to pairs of points; similar structures should then be 'close' in the metric. However, this geometrical construction is not enough for the framing of problems in data-based SHM, as it leaves undefined the notion of feature spaces. Interpreting the feature values on a structure-by-structure basis as a type of field over the space of structures, it seems sensible to borrow an idea from modern theoretical physics, and define feature assignments as sections in a vector bundle over the structure space. With this idea in place, one can interpret the effect of environmental and operational variations as gauge degrees of freedom, as in modern gauge field theories. One can then regard data normalisation procedures like cointegration as gauge-fixing operations. This paper will discuss the various geometrical structures required for an abstract theory of feature spaces in SHM, and will draw analogies with how these structures have shown their power in modern physics. Having motivated a number of problems in Population-Based SHM (PBSHM) in geometrical terms, it remains to show how these problems might be solved. In the second part of the paper, the problem of determining the normal condition cross section of a feature bundle is addressed. The solution is provided by the application of Graph Neural Networks (GNN), a versatile non-Euclidean machine learning algorithm which is not restricted to inputs and outputs from vector spaces. In particular, the algorithm is well suited to operating directly on the sort of graph structures which are an important part of the proposed framework for PBSHM. The solution of the normal section problem is demonstrated for a heterogeneous population of truss structures for which the feature of interest is the first natural frequency. The GNN approach is shown to not only solve the normal section problem, but also to accommodate varying temperatures across the population; it thus provides a means of data normalisation. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the application of a non-linear feature identification technique for structural damage detection. This method is coupled with the impedance-based structural health monitoring (SHM) method, which utilises electromechanical coupling properties of piezoelectric materials. The non-linear feature examined in this study is in the form of autoregressive coefficients in the frequency domain autoregressive model with exogenous (ARX) inputs, which explicitly considers non-linear system input/output relationships. The applicability of this non-linear feature for damage identification is investigated in various frequency ranges using impedance signals measured from a laboratory-test structure. The performance of the non-linear feature is also compared with those of linear features typically used in impedance methods. This paper reinforces the utility of non-linear features in SHM and suggests that their sensitivity in different frequency ranges may be leveraged for certain applications. [Copyright &y& Elsevier]

Abstract: A damage detection method is proposed for structural health monitoring under varying environmental and operational conditions. The method is based on principal component analysis (PCA) applied to vibration features identified during the monitoring of the structure. The advantage of the method is that it does not require to measure environmental parameters because they are taken into account as embedded variables. The number of principal components of the vibration features is implicitly assumed to correspond to the number of independent environmental factors. Since the environmental effects may be effectively eliminated by the proposed procedure, the residual error of the PCA prediction model remains small if the structure is healthy, and it increases significantly when structural damage occurs. Novelty analysis on the residual errors provides a statistical indication of damage. In the present paper, the environmental conditions are assumed to have a linear (or weakly non-linear) effect on the vibration features, and the PCA-based damage detection method is illustrated using computer-simulated and laboratory testing data. The extension of the proposed method to non-linear cases is addressed in a companion paper where the efficiency of the method is verified using data obtained from a 1-year in situ monitoring of a bridge. [Copyright &y& Elsevier]

Abstract: It is well known that changes in vibration features of structures due to damage may be masked by the effects of environmental variations. This influence has to be eliminated in the structural health-monitoring process, especially when a long-term in situ monitoring is expected. In the companion paper [1] a linear method based on principal component analysis (PCA) has been proposed and has shown encouraging results for linear or even weakly non-linear cases. The present paper concerns a further extension of the proposed method to handle non-linear cases, which may be encountered in some complex structures. The method involves a two-step procedure, namely a clustering of the data space into several regions and then the application of PCA in each local region. The application of local PCA allows performing a piecewise linearisation of the non-linear problem. A close look at the choice of the distortion function used in data clustering leads to two new clustering strategies. Whereas the first strategy is specifically suitable for the application treated in this paper, the second one is more general. The local PCA-based damage detection method is applied for the structural health monitoring of a real bridge using vibration data measured in situ over a one-year period. [Copyright &y& Elsevier]

• Contribution of attenuation mechanisms in propagation path can be assessed. • Utilising physics knowledge to inform learning benefits model generation. • Consideration of space over which physics is described can benefit model. The use of ultrasonic guided waves to probe the materials/structures for damage continues to increase in popularity for non-destructive evaluation (NDE) and structural health monitoring (SHM). The use of high-frequency waves such as these offers an advantage over low-frequency methods from their ability to detect damage on a smaller scale. However, in order to assess damage in a structure, and implement any NDE or SHM tool, knowledge of the behaviour of a guided wave throughout the material/structure is important (especially when designing sensor placement for SHM systems). Determining this behaviour is extremely difficult in complex materials, such as fibre–matrix composites, where unique phenomena such as continuous mode conversion takes place. This paper introduces a novel method for modelling the feature-space of guided waves in a composite material. This technique is based on a data-driven model, where prior physical knowledge can be used to create structured machine learning tools; where constraints are applied to provide said structure. The method shown makes use of Gaussian processes, a full Bayesian analysis tool, and in this paper it is shown how physical knowledge of the guided waves can be utilised in modelling using an ML tool. This paper shows that through careful consideration when applying machine learning techniques, more robust models can be generated which offer advantages such as extrapolation ability and physical interpretation. [ABSTRACT FROM AUTHOR]

• A robust fuzzy-driven approach is derived for fully IO/OO automated modal analysis. • The methodology is applicable using any growing model-order identification method. • A new indicator called nMTN is derived for emphasizing weakly excited modes. • The efficiency of the strategy is demonstrated on actual shaking-table tests results. A crucial step when identifying the modal signature of systems using growing order parametric methods consists in discriminating spurious modes from physical modes. In this paper, a three-stages clustering strategy is presented in a fuzzy framework for automating this selection process in the context of Input/Output and Output-Only identification. The novelty and strong point of the approach lies in the first stage where, after computation of single mode validation indicators, a modified fuzzy c-means clustering procedure is developed for performing a first partition. It is shown how the membership function obtained for the cluster of physical modes can be interpreted as a new synthetic modal indicator and helps with pole-splitting detection, outlier rejection and generally improves the final modal parameters estimation. The developed methodology does not involve any user-specified threshold and can be used for discriminating modes produced by any methodology consisting in fitting a growing order model to experimental data of any type. In this paper, accelerations measured during the SMART2013 shaking-table test campaign are processed using data-driven state-space identification algorithms. The automated selection process is used for tracking the modal signature of a trapezoidal shaped reinforced-concrete specimen using in turn stochastic and combined deterministic-stochastic algorithms, defining for the latter the movement of the shaking table as input. Variations in the modal signature are then correlated to the damage actually observed on the specimen and a comparison between Output-Only and Input/Output results is made in order to estimate the interaction between the specimen and the whole shaking table device. [ABSTRACT FROM AUTHOR]

• A particle filter-based structural damage prognosis method using Lamb waves is shown. • A measurement equation is empirically built using numerically simulated Lamb waves. • The method is tested wrt an experimental setup of an Aluminum plate undergoing FCG. • The robustness of the proposed framework over different real specimens is tested. Damage prognosis generally resorts to a damage evolution model and the current damage state to predict the future state and the remaining useful life (RUL). As a properly defined damage-sensitive statistical feature extracted from the Lamb waves can be used to online quantify the damage state, it is here exploited within a particle filter (PF) scheme to perform damage prognosis in structural health monitoring. An accurate mapping between the damage state and this feature would require a sufficient set of experimental Lamb waves collected in correspondence of different damage levels occurring during the run-to-failure process, which, however, is not usually available in real practice due to the high costs and the complex logistics involved in such experimental campaigns. In order to deal with this issue, this paper develops a new numerical simulation-aided particle filter-based damage prognosis framework, where the process equation is still built on the basis of available physical knowledge about the degradation process, whereas the measurement equation is built by means of a data-driven modeling approach using the features extracted from the numerically simulated Lamb waves. The PF framework serves as the state estimation tool which allows to identify the damage state and the parameters in the two equations. The future damage states and the RUL can finally be predicted by projecting the PF estimates in the future using the process equation. The proposed framework is demonstrated with reference to experimental studies of fatigue crack growth in aluminum lug structures with online Lamb wave monitoring. [ABSTRACT FROM AUTHOR]

• A two-dimensional beam-focusing (TDBF) theory is proposed to increase the accuracy and minimize the blind zone of impact localization. • The velocity of acoustic emission signal excited by impact and propagated in plate structure is measured and described as probability distribution. With the TDBF localization algorithm the impact localization result can be displayed with probabilistic tomography which is more precise and intuitive than traditional deterministic description. • The accuracy of the linear sensor array with one dimension beam focus algorithm according to the impact occurred at different angle is measured after multiple tests. To fuse the two tomography calculated using the longitudinal and horizontal sensor array, the Bayes fusion is adopted to fuse the tomography concerning the accuracy which can make the impact localization more precise and credible. Monitoring structural impact is one of the main research aspects of structural health monitoring. A new algorithm for impact localization(IL) is proposed in this paper through a two-dimensional beam-focusing (TDBF) probabilistic tomography theory. The newly developed method is based on the arrangement of cross-shaped piezoelectric sensor (PZT) which has greatly improved the accuracy of IL. Moreover, in the proposed TDBF localization algorithm, the accuracy of the ultrasonic wave velocity achieved during testing has emerged as the main factor that ameliorates the IL. The probability distribution of the velocity is obtained through statistical analysis and introduced into the TDBF localization algorithm to localize the impact through probabilistic tomography. To further improve the accuracy of the IL, Bayes theory is also adopted to fuse two impact tomography obtained by horizontal sensor array and longitudinal sensor array, respectively. The effectiveness and credibility of the TDBF-IL are verified through comprehensive experimental research which indicate the promising prospects of proposed method in accurate and effective monitoring of structural health. [ABSTRACT FROM AUTHOR]

• A mmSHM system is proposed to achieve multi-scale and full-field vibration measurement. • An adaptive segmented phase demodulation method to track vibrations spanning from μm to dm scales. • Full-field vibration measurement method and thoroughly experimental validations. • The accuracy of the mmWave sensing method reaches μm-level. Vibration measurement is of great significance for structural health monitoring (SHM). The emerging microwave vibration measurement technology has distinct advantages over the laser and vision-based approaches, such as long-range, low-power consumption and environmental adaptability. However, it has critical limitations on multi-scale and full-field vibration measurement, especially for SHM of large structures. To overcome these drawbacks, in this paper, we propose mmSHM, a mmWave radar-based vibration measurement approach for SHM. An adaptive segmented phase demodulation (ASPD) method is developed to effectively quantify the displacement variation, allowing accurate displacement measurement spanning from µm to dm scales. What's more, to eliminate the coupling and aliasing interferences from adjacent targets in full field of view, an improved interferometric phase evolution tracking method is presented to extract the vibration displacement from the range-angle joint dimensions. To be specific, the basic principle of vibration measurement using linear frequency-modulated continuous wave (LFMCW) radar is firstly depicted. Then, the detailed procedures of multi-scale and full-field vibration measurement method are illustrated. Finally, with the built prototype of mmSHM, experimental results under different scenarios show that our method can achieve multi-scale and full-field synchronous vibration measurement with µm-level accuracy. The proposed method can be expected to provide a promising approach for contactless vibration measurement in SHM. [ABSTRACT FROM AUTHOR]

• A comprehensive review is presented on the vibration-based signal processing techniques for dynamic feature extraction of structures. • The signal processing techniques is categorized based on their representations in time and frequency domains. • The applications of signal processing techniques are summarized in damage identification of various structures. • The advantages and disadvantages of signal processing techniques are reviewed. Structural health monitoring (SHM) has become an important and hot topic for decades in various fields of civil, mechanical, automotive, and aerospace engineering, etc. Estimating the health condition and understanding the unique characteristics of structures through assessing measured physical parameters in real-time is the major objective of SHM. As a result, signal processing becomes an essential and inseparable approach of vibration based SHM research. The basic goal of using signal processing is to identify the changes or damages from the vibration signals of the dynamic system to detect, locate, and quantify any damages existing in the system. This paper aims to present a comprehensive review of the recent progress that used signal processing techniques for vibration based SHM approaches. Furthermore, the feature extraction process through the signal processing techniques is the basic skeleton of this review. The application of signal processing techniques in structural damage identification procedure is classified into two approaches, namely (i) time-domain and (ii) frequency-domain. Experimental studies have assessed the potentials of the signal processing techniques in two aforementioned domains to enhance the vibration-based structural damage detection subjected to environmental effects. While there have been multiple review studies published on vibration-based structural damage detection, there exists no study in categorizing the signal processing techniques based on the feature extraction procedure that belongs to time and frequency domains for SHM purposes. This review fills this gap and presents a holistic summary of the cutting-edge methodology and technique applied in the relevant research field. [ABSTRACT FROM AUTHOR]

The paper presents the homoscedastic nonlinear cointegration. The method leads to stable variances in nonlinear cointegration residuals. The adapted Breusch–Pagan test procedure is developed to test for the presence of heteroscedasticity (or homoscedasticity) in the cointegration residuals obtained from the nonlinear cointegration analysis. Three different time series – i.e. one with a nonlinear quadratic deterministic trend, simulated vibration data and experimental wind turbine data – are used to illustrate the application of the proposed method. The proposed approach can be used for effective removal of nonlinear trends from various types of data and for reliable structural damage detection based on data that are corrupted by environmental and/or operational nonlinear trends. [ABSTRACT FROM AUTHOR]

This paper investigates a new probabilistic strategy for Bayesian model updating using incomplete modal data. Direct mode matching between the measured and the predicted modal quantities is not required in the updating process, which is realized through model reduction. A Markov chain Monte Carlo technique with adaptive random-walk steps is proposed to draw the samples for model parameter uncertainty quantification. The iterated improved reduced system technique is employed to update the prediction error as well as to calculate the likelihood function in the sampling process. Since modal quantities are used in the model updating, modal identification is first carried out to extract the natural frequencies and mode shapes through the acceleration measurements of the structural system. The proposed algorithm is finally validated by both numerical and experimental examples: a 10-storey building with synthetic data and a 8-storey building with shaking table test data. Results illustrate that the proposed algorithm is effective and robust for parameter uncertainty quantification in probabilistic model updating of buildings. [ABSTRACT FROM AUTHOR]

Randomly missing data of structural vibration responses time history often occurs in structural dynamics and health monitoring. For example, structural vibration responses are often corrupted by outliers or erroneous measurements due to sensor malfunction; in wireless sensing platforms, data loss during wireless communication is a common issue. Besides, to alleviate the wireless data sampling or communication burden, certain accounts of data are often discarded during sampling or before transmission. In these and other applications, recovery of the randomly missing structural vibration responses from the available, incomplete data, is essential for system identification and structural health monitoring; it is an ill-posed inverse problem, however. This paper explicitly harnesses the data structure itself—of the structural vibration responses—to address this (inverse) problem. What is relevant is an empirical, but often practically true, observation, that is, typically there are only few modes active in the structural vibration responses; hence a sparse representation (in frequency domain) of the single-channel data vector, or, a low-rank structure (by singular value decomposition) of the multi-channel data matrix. Exploiting such prior knowledge of data structure (intra-channel sparse or inter-channel low-rank), the new theories of ℓ 1 -minimization sparse recovery and nuclear-norm-minimization low-rank matrix completion enable recovery of the randomly missing or corrupted structural vibration response data. The performance of these two alternatives, in terms of recovery accuracy and computational time under different data missing rates, is investigated on a few structural vibration response data sets—the seismic responses of the super high-rise Canton Tower and the structural health monitoring accelerations of a real large-scale cable-stayed bridge. Encouraging results are obtained and the applicability and limitation of the presented methods are discussed. [ABSTRACT FROM AUTHOR]

An easy to use, fast to apply, cost-effective, and very accurate non-destructive testing (NDT) technique for damage localisation in complex structures is key for the uptake of structural health monitoring systems (SHM). Acoustic emission (AE) is a viable technique that can be used for SHM and one of the most attractive features is the ability to locate AE sources. The time of arrival (TOA) technique is traditionally used to locate AE sources, and relies on the assumption of constant wave speed within the material and uninterrupted propagation path between the source and the sensor. In complex structural geometries and complex materials such as composites, this assumption is no longer valid. Delta T mapping was developed in Cardiff in order to overcome these limitations; this technique uses artificial sources on an area of interest to create training maps. These are used to locate subsequent AE sources. However operator expertise is required to select the best data from the training maps and to choose the correct parameter to locate the sources, which can be a time consuming process. This paper presents a new and improved fully automatic delta T mapping technique where a clustering algorithm is used to automatically identify and select the highly correlated events at each grid point whilst the “Minimum Difference” approach is used to determine the source location. This removes the requirement for operator expertise, saving time and preventing human errors. A thorough assessment is conducted to evaluate the performance and the robustness of the new technique. In the initial test, the results showed excellent reduction in running time as well as improved accuracy of locating AE sources, as a result of the automatic selection of the training data. Furthermore, because the process is performed automatically, this is now a very simple and reliable technique due to the prevention of the potential source of error related to manual manipulation. [ABSTRACT FROM AUTHOR]

The paper describes a kinematic model-based solution to estimate simultaneously the calibration parameters of the vision system and the full-motion (6-DOF) of large civil engineering structures, namely of long deck suspension bridges, from a sequence of stereo images captured by digital cameras. Using an arbitrary number of images and assuming a smooth structure motion, an Iterated Extended Kalman Filter is used to recursively estimate the projection matrices of the cameras and the structure full-motion (displacement and rotation) over time, helping to meet the structure health monitoring fulfilment. Results related to the performance evaluation, obtained by numerical simulation and with real experiments, are reported. The real experiments were carried out in indoor and outdoor environment using a reduced structure model to impose controlled motions. In both cases, the results obtained with a minimum setup comprising only two cameras and four non-coplanar tracking points, showed a high accuracy results for on-line camera calibration and structure full motion estimation. [ABSTRACT FROM AUTHOR]

Catastrophic structural failure of aircrafts, bridges, buildings and other structures in modern societies has always been of primary concern because of the loss of human lifes and of negative economic impact. The aging of the structures, the growing dependency on their role in our networks of transportation, energy and comunications, the smaller construction tolerances, the bigger power demanded and the media and society awardness to catastrophic events are sufficient motivations for the growing field of structural health monitoring, which aims at assessing the actual condition of a structure and to identify incipient damage. Damage identification can be considered as a two step process, the detection and the diagnosis. The former, and fundamental step, is the confirmation of an efective damage existence. When the response is affirmative, the latter step begins with the diagnosis, and then the questions are: where?, how much?, what type?, when will it fail? In this paper the authors propose a simple method to detect and relatively quantify structural damage by using measured vibrations data, specifically the operational deflections shapes. Numerical simulations and experimental tests are presented to validate the proposed method. [ABSTRACT FROM AUTHOR]

Structural health monitoring of wind turbines is usually performed by collecting real-time operating data on a limited number of accessible locations using traditional sensors such as accelerometers and strain-gauges. When dealing with offshore wind turbines (OWT) though, most of the fatigue sensitive spots are inaccessible for direct measurements, e.g. at the mudline below the water level. Response estimation techniques can then be used to estimate the response at unmeasured locations from a limited set of response measurements and a Finite Element Model. In this paper the method will be validated using accelerations only. The method makes use of a modal decomposition and expansion algorithm that allows for successful response prediction. The algorithm is first validated using simulated datasets provided from HAWC2 and then using real time data obtained from a monitoring campaign on an offshore Vestas V90 3 MW wind turbine on a monopile foundation in the Belgian North Sea. [ABSTRACT FROM AUTHOR]

Contrary to global modal parameters such as eigenfrequencies, mode shapes inherently provide structural information on a local level. Therefore, this particular modal parameter and its derivatives are utilized extensively for damage identification. Typically, more or less advanced mathematical methods are employed to identify damage-induced discontinuities in the spatial mode shape signals, hereby, potentially, facilitating damage detection and/or localization. However, by being based on distinguishing damage-induced discontinuities from other signal irregularities, an intrinsic deficiency in these methods is the high sensitivity towards measurement noise. In the present paper, a damage localization method which, compared to the conventional mode shape-based methods, has greatly enhanced robustness towards measurement noise is proposed. The method is based on signal processing of a spatial mode shape by means of continuous wavelet transformation (CWT) and subsequent application of a generalized discrete Teager–Kaiser energy operator (GDTKEO) to identify damage-induced mode shape discontinuities. In order to evaluate whether the identified discontinuities are in fact damage-induced, outlier analysis is conducted by applying the Mahalanobis metric to major principal scores of the sensor-located bands of the signal-processed mode shape. The method is tested analytically and benchmarked with other mode shape-based damage localization approaches on the basis of a free-vibrating beam and validated experimentally in the context of a residential-sized wind turbine blade subjected to an impulse load. [ABSTRACT FROM AUTHOR]

The stiffness characteristics of the contact interfaces in joints or boundary conditions have a great effect on dynamic response of assembled structures. Predictive analytical/numerical modeling of mechanical structures is not possible without representing the contact interfaces accurately. Because of the complex mechanisms involved, contact interfaces introduce difficulties both in modeling the inherent dynamics and identification of the model parameters. In this paper an identification approach employing the dynamic characteristic equation is proposed for linear interface parameters. The proposed method is applicable to both analytical and numerical problems. The accuracy of the proposed method is investigated by simulation results of a beam with elastic boundary support and experimental results of a bolted lap-joint. [ABSTRACT FROM AUTHOR]

Over the last three decades, there have been increasing demands to develop and deploy Structural Health Monitoring (SHM) systems for engineering structures in service. Since these structures are subjected to varying environmental and operational conditions, reliable SHM methodologies must be capable of not misattributing to damage changes due to environmental conditions. This paper presents a novel damage detection methodology based on the similarity between maxima decay lines of the continuous wavelet transform scalogram of the structural responses obtained under different operational and environmental conditions. The normalized cross correlation (NCC) is used as a measure of this similarity. In addition, the pointwise summation of similar Wavelet Transform Modulus Maxima (WTMM) decay lines is used to identify changes due to the presence of damage from different force realizations and/or varying environmental conditions. The effectiveness of the proposed methodology is demonstrated using a simulated 3DOF system and an experimental cantilever beam. [ABSTRACT FROM AUTHOR]

This paper is concerned with the development of a general methodology for identifying mechanical sources from prior local information on both their nature and location over the studied structure. For this purpose, the formulation of the identification problem is derived from the Bayesian statistics, that provides a flexible way to account for local a priori on the distribution of sources. Practically, the resulting optimization problem can be seen as a group generalized Tikhonov regularization, that is solved in an iterative manner. The main features of the proposed identification method are illustrated with both numerical and experimental examples. In particular, it is shown that properly exploiting the local spatial information drastically improves the quality of the source identification. [ABSTRACT FROM AUTHOR]

In this paper, results of damage localization performed for four sensing network configurations are compared. Process of damage localization is based on guided waves propagation phenomenon. Guided waves are excited using piezoelectric transducer and received by scanning laser vibrometer. Different excitation frequencies are also investigated. In experimental investigations two types of piezoelectric transducers are used as guided waves exciters. Frequency–magnitude characteristics of symmetric and antisymmetric modes are created for both types of transducers. These characteristics allow a choice of an excitation frequency for efficient generation of selected wave mode. The amplitude of second mode in this case has negligibly small value. Finally, sensing networks in the form of circle with three different diameters are realized based on piezoelectric transducers. Damage localization algorithm is prepared in MATLAB ® environment as well as in C++. [ABSTRACT FROM AUTHOR]

In the last decades the need for an effective seismic protection and vulnerability reduction of cultural heritage buildings and sites determined a growing interest in structural health monitoring (SHM) as a knowledge-based assessment tool to quantify and reduce uncertainties regarding their structural performance. Monitoring can be successfully implemented in some cases as an alternative to interventions or to control the medium- and long-term effectiveness of already applied strengthening solutions. The research group at the University of Padua, in collaboration with public administrations, has recently installed several SHM systems on heritage structures. The paper reports the application of monitoring strategies implemented to avoid (or at least minimize) the execution of strengthening interventions/repairs and control the response as long as a clear worsening or damaging process is detected. Two emblematic case studies are presented and discussed: the Roman Amphitheatre (Arena) of Verona and the Conegliano Cathedral. Both are excellent examples of on-going monitoring activities, performed through static and dynamic approaches in combination with automated procedures to extract meaningful structural features from collected data. In parallel to the application of innovative monitoring techniques, statistical models and data processing algorithms have been developed and applied in order to reduce uncertainties and exploit monitoring results for an effective assessment and protection of historical constructions. Processing software for SHM was implemented to perform the continuous real time treatment of static data and the identification of modal parameters based on the structural response to ambient vibrations. Statistical models were also developed to filter out the environmental effects and thermal cycles from the extracted features. [ABSTRACT FROM AUTHOR]

In this paper, a probabilistic multi-class pattern recognition algorithm is developed for damage monitoring of smart structures. As these structures can face damages of different severities located in various positions, multi-class classifiers are needed. We propose an original support vector machine (SVM) multi-class clustering algorithm that is based on a probabilistic decision tree (PDT) that produces a posteriori probabilities associated with damage existence, location and severity. The PDT is built by iteratively subdividing the surface and thus takes into account the structure geometry. The proposed algorithm is very appealing as it combines both the computational efficiency of tree architectures and the SVMs classification accuracy. Damage sensitive features are computed using an active approach based on the permanent emission of non-resonant Lamb waves into the structure and on the recognition of amplitude disturbed diffraction patterns. The effectiveness of this algorithm is illustrated experimentally on a composite plate instrumented with piezoelectric elements. [ABSTRACT FROM AUTHOR]

The study described in this paper revisits the concept of instantaneous identification of system parameters based on time–frequency representations. In order to overcome the distortion caused by the time–frequency analysis window, optimal bias-compensated estimators are introduced. In particular, bias-compensated estimators are conceived specifically to provide on-line estimates for time-varying linear parameters such as instantaneous frequency and damping. With respect to previous studies, which relied on the concept of optimal time–frequency representation, the novel procedure corrects on-line estimates provided by standard representations. Afterwards, a practical application to a full-scale structure is presented. The church of “Madonnina della Neve” in Savigliano (Cuneo province, Italy) is a masonry building that exhibits defects in the connections between structural parts and visible cracks in lateral masonry walls. As a consequence, the global behaviour observed on this structure demonstrates significant flexibility in both the longitudinal and transversal directions. Recently cracks have worsened due to vibrations induced by traffic and heavy vehicles. Proposed bias-compensated estimators were used to analyse the unusual time-varying behaviour of the masonry structure, as observed from the response measured during the transit of vehicles and trucks. [ABSTRACT FROM AUTHOR]

This paper investigates the use of relative displacement measurements from the newly developed relative displacement sensors to identify the damage of shear connectors in composite bridges. Continuous Wavelet Transform and Hilbert–Huang Transform are applied to analyze the measured dynamic responses and to identify the damage of shear connectors in the composite bridge model under moving loads. Comparative studies by using the relative displacement, acceleration and displacement measurements respectively for the damage detection are conducted. A comparative study of using relative displacements and acceleration responses of the bridge under ambient excitations to monitor the shear connector conditions is also conducted. Numerical and experimental studies demonstrate that both relative displacement and acceleration measurements can identify the location and the instant of damage occurrence in shear connectors when the bridge is under moving loads. The results demonstrate that relative displacement is a better response quantity for structural health monitoring of composite bridges. [ABSTRACT FROM AUTHOR]

• This paper presents a comprehensive review of the general architecture of 1D CNNs. • Their major engineering applications, principals, and recent progress on 1D CNNs are reviewed. • The state-of-the-art performance and unique properties of 1D CNNs are highlighted. • Detailed computational complexity analysis of compact and adaptive 1D CNNs are reported. • The benchmark datasets and the principal 1D CNN software are also publicly shared. During the last decade, Convolutional Neural Networks (CNNs) have become the de facto standard for various Computer Vision and Machine Learning operations. CNNs are feed-forward Artificial Neural Networks (ANNs) with alternating convolutional and subsampling layers. Deep 2D CNNs with many hidden layers and millions of parameters have the ability to learn complex objects and patterns providing that they can be trained on a massive size visual database with ground-truth labels. With a proper training, this unique ability makes them the primary tool for various engineering applications for 2D signals such as images and video frames. Yet, this may not be a viable option in numerous applications over 1D signals especially when the training data is scarce or application specific. To address this issue, 1D CNNs have recently been proposed and immediately achieved the state-of-the-art performance levels in several applications such as personalized biomedical data classification and early diagnosis, structural health monitoring, anomaly detection and identification in power electronics and electrical motor fault detection. Another major advantage is that a real-time and low-cost hardware implementation is feasible due to the simple and compact configuration of 1D CNNs that perform only 1D convolutions (scalar multiplications and additions). This paper presents a comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field. Their state-of-the-art performance is highlighted concluding with their unique properties. The benchmark datasets and the principal 1D CNN software used in those applications are also publicly shared in a dedicated website. While there has not been a paper on the review of 1D CNNs and its applications in the literature, this paper fulfills this gap. [ABSTRACT FROM AUTHOR]

• For the first time, underwater multi-bolt looseness detection is investigated. • Two new entropy indexes are proposed, including M RangeEn and M bEn. • A new entropy-enhanced active sensing method is developed. • A stacking-based ensemble learning classifier is trained to detect bolt looseness. • A lab-level experiment is conducted to verify the proposed method. With the rapid development of the oil industry, more and more subsea pipelines enter into service, and thus we are facing the challenge of safe operation and maintenance of these underwater infrastructures. For instance, as a common component to connect subsea pipelines, the flange is prone to losing sufficient integrity due to severe working conditions and harsh environments under the sea. However, most of the current structural health monitoring (SHM) and nondestructive testing (NDT) techniques are targeted at onshore structures, and we require a more practical method for multi-bolt looseness identification of subsea flanges. In this paper, attempting to solve the above difficulty, we propose the "Smart Crawfish", which is a concept that employs a subsea robot equipped with a gripper and a pair of Lead Zirconate Titanate (PZT) transducers to implement the active sensing method. Particularly, the main computational contribution of this paper is that we develop two new entropy indexes, i.e., multiscale range entropy (M RangeEn) and multiscale bubble entropy (M bEn) to enhance the current active sensing method (i.e., the entropy-enhanced active sensing method). To detect underwater multi-bolt looseness, we first obtain stress wave signals under different categories (i.e., different cases of multi-bolt looseness) by the active sensing and then compute their M RangeEn and M bEn to construct the feature dataset. Subsequently, via the dataset, a stacking-based ensemble learning classifier is trained to classify different categories. Finally, we conduct a lab-level experiment to demonstrate the effectiveness of the proposed method, which outperforms the current entropy-enhanced active sensing. To the best of our knowledge, this paper is the first attempt to address the detection of underwater multi-bolt looseness, which has great potential for future industrial applications. [ABSTRACT FROM AUTHOR]

• For the first time a new population-based methodology is proposed that alleviates problems with sparsity of data in SHM. • The applicability of domain adaptation assumptions when considering each population type and SHM problem are outlined. • A novel MMD-based algorithm is proposed for an (L + 1)-problem. • A new approach for using physics-based models and domain adaption is proposed for labelling data from structures in PBSHM. • This is the third paper (of three parts) covering the foundations of population-based SHM. This is the third and final paper in a series laying foundations for a theory/methodology of Population-Based Structural Health Monitoring (PBSHM). PBSHM involves utilising knowledge from one set of structures in a population and applying it to a different set, such that predictions about the health states of each member in the population can be performed and improved. Central ideas behind PBSHM are those of knowledge transfer and mapping. In the context of PBSHM, knowledge transfer involves using information from a source domain structure, where labels are known for given feature sets, and mapping these onto the unlabelled feature space of a different, target domain structure. This mapping means a classifier trained on the transformed source domain data will generalise to the unlabelled target domain data; i.e. a classifier built on one structure will generalise to another, making Structural Heath Monitoring (SHM) cost-effective and applicable to a wide range of challenging industrial scenarios. This process of mapping features and labels across source and target domains is defined here via domain adaptation , a subcategory of transfer learning. A mathematical underpinning for when domain adaptation is possible in a structural dynamics context is provided, with reference to topology within a graphical representation of structures. Subsequently, a novel procedure for performing domain adaptation on topologically different structures is outlined. [ABSTRACT FROM AUTHOR]

• Second paper in a three-part series laying a foundation for population-based SHM (PBSHM). • Introduces the Irreducible Element (IE) model as a basis for a novel abstract representation of structures. • Introduces the Attributed Graph (AG) model as a means of measuring degree of similarity between structures. • Provides formal definitions for the objects in the PBSHM theory. This paper is the second in a series of three which aims to provide a basis for Population-Based Structural Health Monitoring (PBSHM); a new technology which will allow transfer of diagnostic information across a population of structures, augmenting SHM capability beyond that applicable to individual structures. The new PBSHM can potentially allow knowledge about normal operating conditions, damage states, and even physics-based models to be transferred between structures. The first part in this series considered homogeneous populations of nominally-identical structures. The theory is extended in this paper to heterogeneous populations of disparate structures. In order to achieve this aim, the paper introduces an abstract representation of structures based on Irreducible Element (IE) models, which capture essential structural characteristics, which are then converted into Attributed Graphs (AGs). The AGs form a complex network of structure models, on which a metric can be used to assess structural similarity; the similarity being a key measure of whether diagnostic information can be successfully transferred. Once a pairwise similarity metric has been established on the network of structures, similar structures are clustered to form communities. Within these communities, it is assumed that a certain level of knowledge transfer is possible. The transfer itself will be accomplished using machine learning methods which will be discussed in the third part of this series. The ideas introduced in this paper can be used to define precise terminology for PBSHM in both the homogeneous and heterogeneous population cases. [ABSTRACT FROM AUTHOR]

The maintenance and repair of bridges (and other large scale infrastructure projects) is a major area which could benefit from Structural Health Monitoring technology. Inspections on bridges can take a long time and require many people, and are therefore conducted infrequently. This low frequency of inspection leaves the chance that damage and dangerous critical failures can occur during the long timeframes between inspections. It might even be the case that an inspection fails to identify sub-surface damage. Therefore some form of continuous monitoring is desirable, especially if such systems can reliably detect sub-surface damage. However, the application of SHM to bridges is made challenging by the cost and practicability of obtaining damage-state data for bridges. Over the lifetime of a single bridge, it is hoped that a critical failure will never occur, and only a small number of the possible damage states will occur. It is also unpractical to intentionally damage structures to obtain damage-state data. Population-based structural health monitoring seeks to overcome the obstacle of the limited data available for a single structure, by allowing data to be shared between similar structures. Bridges represent an interesting challenge for PBSHM as each bridge is unique. As such, an assessment of how similar bridges are to each other is required. To provide this assessment, one must develop an abstract representation for each bridge, and using this to perform a comparison. This paper describes the use of a general approach for assessing the similarity of structures, applied to several bridge examples which are representative of common types of bridges, to show that it can be applied in this field. [ABSTRACT FROM AUTHOR]

• The matching-combination method of vibration modes is developed to construct the combined vibration mode. • The combined modal strain energy index is proposed based on the difference between the original mode and the combined mode. • The results of numerical examples show that the proposed method features high damage localization accuracy. The existing damage detection methods are challenging to be directly applied to single-layer latticed shells due to the complex mechanical mechanism and modal characteristics of such structures. In this regard, an effective damage detection method based on the combined modal strain energy index is proposed in this paper to detect the location of the damaged members in such structures. Based on the vibration modal features before and after damage of structures, the modal matching and combination method is presented to generate the combined mode, which include the modal matching procedure based on the modal assurance criterion and the modal combination procedure using the Kriging interpolation. Compared with the damaged mode, the specially constructed combined mode can significantly reduce the mode shape discrepancy in the undamaged region on the basis of retaining the characteristic differences in the damaged members. On this basis, the combined modal strain energy change index is proposed according to the modal strain energy difference between the original mode and the combined mode. Since the proposed index is established according to the difference between the combined vibration mode and the original vibration mode, the values of damage index of the undamaged members are greatly reduced, so as to effectively avoid the occurrence of misjudgement. Several numerical examples with different damage scenarios are studied in order to evaluate the proposed approach. The results clearly indicate that the proposed method features high damage localization accuracy compared with traditional methods and good measurement noise robustness. [ABSTRACT FROM AUTHOR]

• The DDL is decomposed by the hierarchical-clustering-oriented RBF framework at each time step. • The multi-dimensional interval model is used to quantify the convex-fuzzy hybrid uncertainties. • The fuzzy interval bounds of the DDL can be obtained based on the Chebyshev-interval surrogate model. Load identification is a hotly studied topic due to the widespread recognition of its importance in structural design and health monitoring. This paper explores an effective identification method for the distributed dynamic load (DDL) varying in both time progress and space dimensions using limited acceleration responses. As for the reconstruction of spatial distribution, the radial basis function (RBF) interpolation strategy, whose hyper-parameters are determined by a hierarchical clustering algorithm, is applied to approximate the DDL and then transform the continuous function into finite dimensions. In the time domain, based on the inverse Newmark iteration, the RBF coefficients at each discrete instant are obtained by the least square solution of the modal forces. Considering the multi-source uncertainties lacking exact probability distributions, a multi-dimensional interval model is developed to quantify convex parameters and fuzzy parameters uniformly. Further, a Chebyshev-interval surrogate model with different orders is constructed to obtain the fuzzy-interval boundaries of DDLs. Eventually, three examples are discussed to demonstrate the feasibility of the developed DDL identification approach considering hybrid uncertainties. The results suggest its promising applications in different structures and loading conditions. [ABSTRACT FROM AUTHOR]

Population-based structural health monitoring (PBSHM) offers a new viewpoint for structural health monitoring (SHM), allowing diagnostic information to be shared across populations of structures. By extending the set of available damage observations, a population-based approach can diagnose damage previously unseen on a structure of interest by leveraging damage information from other structures in the population. These technologies therefore provide significant benefits for making SHM practicable in a variety of industrial settings. It is proposed that PBSHM methodologies must be comprised of tools to assess similarity coupled with algorithms that perform knowledge transfer. The similarity tools are important for identifying whether an SHM task should be attempted for a given population by assessing both the structural similarities between members of a population and similarities in their data spaces. An abstract representation of structures in a graphical domain is presented as an objective way of assessing structural similarity, with distance metrics utilised for assessing data-space similarities. Knowledge transfer is performed using a branch of transfer learning called domain adaptation. By determining if members of a population are similar in a structural and data-space sense, the risk of negative transfer can be reduced; whereby domain adaptation reduces classification performance. This paper demonstrates a PBSHM methodology for transferring knowledge within a heterogeneous population (a group of non-identical structures). Specifically, the PBSHM methodology is shown to transfer localisation labels from a Gnat aircraft wing to an unlabelled Piper Tomahawk aircraft wing dataset, resulting in 100% classification accuracy. [Display omitted] • A population-based SHM methodology is presented for heterogeneous populations. • Methodology quantifies structural and data similarity before knowledge transfer. • Abstract representations of structures as graphs quantify structural similarity. • Domain adaptation transfers label knowledge between structures. • The methodology is demonstrated on a population of two experimental aircraft wings. [ABSTRACT FROM AUTHOR]

Over the past decade, the ideologies surrounding Structural Health Monitoring (SHM) have shifted drastically within the aerospace engineering disciplines, predominantly onus to rapid advancements in machine intelligence. While traditional SHM practices are based on scheduled and pre-emptive maintenance, the NextGen SHM system, known commonly as Prognostics and Health Management (PHM), has a focus on pro-active condition-based maintenance, forecasting and prognostics — a milestone on the trajectory towards Digital Twin technology. In aircraft, particularly defense fighter air platforms, fatigue-critical high-amplitude cyclic behavior is unavoidable, where rapid fatigue life consumption due to an airframe buffet is one of the most problematic phenomena that engineers have encountered throughout the 4th and 5th generation fighter programs. This paper serves as a point-of-reference consolidating a range of machine learning models, under a single benchmark aircraft Multi-Input Single-Output (MISO) loads monitoring problem. Linear regression models, traditional (shallow) artificial neural networks, and deep learning strategies are all explored, where strain sensors are used as inputs to predict representative bending and torsional dynamic (buffet) and quasi-static (maneuver) load spectra on an aircraft wing during transonic buffeting maneuvers. For the benchmark system considered herein, the MISO coherence ranges from high to very weak depending on the load case, hereby providing a unique opportunity to rigorously explore the time-series modeling requirements and make valuable recommendations across a wide range of data-qualities that are likely to be encountered in traditional or modern aircraft data-acquisition systems or, for that matter, in any mechanical systems plagued by fatigue. [ABSTRACT FROM AUTHOR]

• A recursive window-based Bayesian inference approach is formulated and validated. • It is implemented for joint input-parameter estimation of an offshore structure. • Strain time histories at hotspots are accurately predicted through virtual sensing. • The proposed method outperform modal expansion method for virtual sensing. This paper presents a recursive Bayesian inference framework for joint parameter-input identification, and virtual sensing for strain time history prediction of an offshore platform using sparse output-only measurements. The studied offshore platform, known as FINO3, is in the North Sea and is instrumented with a variety of sensors, including accelerometers and strain gauges. Offshore platforms are fatigue critical structures due to harsh marine environmental conditions and continuous cyclic wind and wave loads. Therefore, continuous monitoring of strain time histories at hotspot locations of offshore structures is important for reducing maintenance cost and avoiding unexpected failures. A windowed unscented Kalman filter (UKF) is employed to estimate an uncertain modeling parameter (foundation stiffness) and unknown input load time histories using output-only acceleration and strain measurements. The input loads are divided into overlapping windows, and windowed inputs and model parameters are combined as an augmented state vector in the UKF framework. Then strain time histories at critical locations are estimated through a virtual sensing strategy using the estimated input loads and model parameter. A traditional modal expansion approach combined with model updating is also implemented for the purpose of verification and comparison. The proposed method is first demonstrated through a numerical study using a finite element model of FINO3, where accurate model parameter and input estimations are obtained. Then the approach is further investigated using the actual measurements on FINO3. More accurate strain predictions are provided by the UKF than the modal expansion approach, which recommends the proposed UKF method for fatigue monitoring and input estimation. [ABSTRACT FROM AUTHOR]

Damage diagnosis based on global structural vibrations critically depends on the sensor layout, in particular when a small number of sensors is used for large structures under unknown excitation. This paper proposes a sensor placement strategy that yields an optimized sensor layout with maximum damage detectability in selected structural components. The optimization criterion is based on the Fisher information, which quantifies the information that the damage-sensitive feature carries on the design parameters of structural components, such as material constants or cross-sectional values. It is evaluated using a finite element model, and considers the statistical uncertainties of the damage-sensitive feature. The methodology is shown for the stochastic subspace-based damage detection method, but can be applied to any damage-sensitive feature whose distribution can be approximated as Gaussian. It is suitable to find the optimal layout for a fixed number of sensors and to choose an appropriate number of sensors. Since the Fisher information is defined component-wise, the sensor layout can be tuned to become more sensitive to damage in local structural components, such as damage hotspots, non-inspectable components, or components that are critical for the safety and serviceability of the structure. For proof of concept, the sensor layout on a laboratory beam is optimized based on numerical simulations, and it is showcased that the optimal sensor layout leads to the highest damage detectability for experimental data. • Gaussian vibration-based features are employed for statistical damage detection. • The sensor layout is optimized to maximize the feature's sensitivity and minimize its uncertainty. • The optimization criterion allows one to tune the sensor layout for high detectability in critical components. • A custom-made genetic algorithm efficiently finds a close-to-optimal solution for large structures. • The optimization is based on a numerical model and validated in a lab experiment. [ABSTRACT FROM AUTHOR]

Acoustic emission (AE) has been shown to be a promising health monitoring technique for composite materials as it allows real-time location and identification of damage. When attempting to relate the recorded acoustic emission to a material׳s mechanical behavior, the relevance of results relies on an accurate selection of AE originating from material damage. Indeed, during mechanical tests most of the recorded AE is generated outside of the volume of interest and without proper filtering these AE signals can significantly affect the analysis. To date, there exists no common procedure for the selection of AE signals and therefore results can hardly be compared or transposed to different composite systems. In the present paper, a new waveform-based procedure is proposed for the selection of AE events generated by damage. The procedure includes accurate localization and selection assessment based on signal energy. [ABSTRACT FROM AUTHOR]

This paper aims to compare several damage identification methods based on the analysis of modal curvature and related quantities (natural frequencies and modal strain energy) by evaluating their performances on the same test case, a damaged Euler–Bernoulli beam. Damage is modelled as a localized and uniform reduction of stiffness so that closed-form expressions of the mode-shape curvatures can be analytically computed and data accuracy, which affects final results, can be controlled. The selected techniques belong to two categories: one includes several methods that need reference data for detecting structural modifications due to damage, the second group, including the modified Laplacian operator and the fractal dimension, avoids the knowledge of the undamaged behavior for issuing a damage diagnosis. To explain better the different performances of the methods, the mathematical formulation has been revised in some cases so as to fit into a common framework where the underlying hypotheses are clearly stated. Because the various damage indexes are calculated on ‘exact’ data, a sensitivity analysis has been carried out with respect to the number of points where curvature information is available, to the position of damage between adjacent points, to the modes involved in the index computation. In this way, this analysis intends to point out comparatively the capability of locating and estimating damage of each method along with some critical issues already present with noiseless data. [ABSTRACT FROM AUTHOR]

The use of condition monitoring (CM) data to predict remaining useful life have been growing with increasing use of health and usage monitoring systems on aircraft. There are many data-driven methodologies available for the prediction and popular ones include artificial intelligence and statistical based approach. The drawback of such approaches is that they require a lot of failure data for training which can be scarce in practice. In lieu of this, methods using state-space and regression-based models that extract information from the data history itself have been explored. However, such methods have their own limitations as they utilize a single time-invariant model which does not represent changing degradation path well. This causes most degradation modeling studies to focus only on segments of their CM data that behaves close to the assumed model. In this paper, a state-space based method; the Switching Kalman Filter (SKF), is adopted for model estimation and life prediction. The SKF approach however, uses multiple models from which the most probable model is inferred from the CM data using Bayesian estimation before it is applied for prediction. At the same time, the inference of the degradation model itself can provide maintainers with more information for their planning. This SKF approach is demonstrated with a case study on gearbox bearings that were found defective from the Republic of Singapore Air Force AH64D helicopter. The use of in-service CM data allows the approach to be applied in a practical scenario and results showed that the developed SKF approach is a promising tool to support maintenance decision-making. [ABSTRACT FROM AUTHOR]

A comprehensive comparison of macro-strain mode and displacement mode obtained from distributed macro-strain sensing and high-density point sensing (such as accelerometers) technologies is presented in this paper. Theoretical derivation reveals that displacement mode shape from accelerometers and modal macro-strain from distributed macro-strain sensors can be converted into each other. However, it is realized that displacement mode shape as global behavior of a structure can still be calculated with high-precision from modal macro-strain considering measurement errors in practical monitoring, whereas modal macro-strain can hardly be accurately achieved from displacement mode shape when signals are corrupted with noise in practical monitoring. Simulation and experiment results show that the calculated displacement mode shapes are very close to the actual ones even if the noise level reaches 5%. Meanwhile, damage index using measured modal macro-strain is still effective when the measurements are corrupted with 5% noise which is reliable for damage detection in practical monitoring. Calculating modal macro-strain from noise-polluted displacement mode shape will cause an unacceptable error if the noise level reaches only 0.5%, which has been verified in the simulation. [ABSTRACT FROM AUTHOR]

System health diagnostics provides diversified benefits such as improved safety, improved reliability and reduced costs for the operation and maintenance of engineered systems. Successful health diagnostics requires the knowledge of system failures. However, with an increasing system complexity, it is extraordinarily difficult to have a well-tested system so that all potential faulty states can be realized and studied at product testing stage. Thus, real time health diagnostics requires automatic detection of unexampled system faulty states based upon sensory data to avoid sudden catastrophic system failures. This paper presents a trifold hybrid classification (THC) approach for structural health diagnosis with unexampled health states (UHS), which comprises of preliminary UHS identification using a new thresholded Mahalanobis distance (TMD) classifier, UHS diagnostics using a two-class support vector machine (SVM) classifier, and exampled health states diagnostics using a multi-class SVM classifier. The proposed THC approach, which takes the advantages of both TMD and SVM-based classification techniques, is able to identify and isolate the unexampled faulty states through interactively detecting the deviation of sensory data from the exampled health states and forming new ones autonomously. The proposed THC approach is further extended to a generic framework for health diagnostics problems with unexampled faulty states and demonstrated with health diagnostics case studies for power transformers and rolling bearings. [ABSTRACT FROM AUTHOR]

Abstract: For structural health monitoring applications there is a need for simple and contact-less methods of Non-Destructive Evaluation (NDE). A number of damage detection techniques have been developed, such as frequency shift, generalised fractal dimension and wavelet transforms with the aim to identify, locate and determine the severity of damage in a material or structure. These techniques are often tailored for factors such as (i) type of material, (ii) damage pattern (crack, delamination), and (iii) the nature of any input signals (space and time). This paper describes and evaluates a wavelet-based damage detection framework that locates damage on cantilevered beams via NDE using computer vision technologies. The novelty of the approach is the use of computer vision algorithms for the contact-less acquisition of modal shapes. Using the proposed method, the modal shapes of cantilever beams are reconstructed by extracting markers using sub-pixel Hough Transforms from images captured using conventional slow motion cameras. The extracted modal shapes are then used as an input for wavelet transform damage detection, exploiting both discrete and continuous variants. The experimental results are verified and compared against finite element analysis. The methodology enables a non-invasive damage detection system that avoids the need for expensive equipment or the attachment of sensors to the structure. Two types of damage are investigated in our experiments: (i) defects induced by removing material to reduce the stiffness of a steel beam and (ii) delaminations in a composite laminate. Results show successful detection of notch depths of 5%, 28% and 50% for the steel beam and of 30mm delaminations in central and outer layers for the composite laminate. [Copyright &y& Elsevier]

Abstract: Cointegration has now been established as a powerful means of projecting out long-term trends from time-series data in the context of econometrics. Recent work has further established that cointegration can be applied profitably in the context of structural health monitoring (SHM), where it is desirable to project out the effects of environmental and operational variations from data in order that they do not generate false positives in diagnostic tests. These variations typically impose a time-varying nature on SHM data on a much longer time scale than those at which the dynamics of the structures of interest usually manifest themselves. This situation means that cointegration can be used to remove long-term time variations and thus projects out the operational and environmental effects. The concept of cointegration is partly built on a clear understanding of the ideas of stationarity and nonstationarity for time-series. The current paper will discuss the distinction between different types of nonstationarity in the context of SHM data and will extend the discussion by the introduction of multiresolution (discrete wavelet) analysis as a means of characterising the time scales on which nonstationarity manifests itself. One of the major results of the current study is to demonstrate how a multiresolution approach to cointegration can lead to enhanced sensitivity for damage detection. The discussion will be based on synthetic data and also on experimental data for the guided-wave SHM of a composite plate. [Copyright &y& Elsevier]

Abstract: This paper proposes a novel output-only damage identification method based on the unsupervised blind source separation (BSS) technique termed independent component analysis (ICA). It is discovered that ICA biases to extract sparse component, which typically indicates damage, from the observed mixture signals. The measured structural responses are first preprocessed by wavelet transform (WT). The wavelet-domain signals are then fed as mixtures into the BSS model, which is solved by ICA. The obtained “interesting” source with sharp spike and its associated spatial signature in the recovered mixing matrix reveal damage instant and location respectively. Following which, identification of the time-varying modes is carried out by ICA using the structural responses before and after the identified damage instant. For illustration, numerical simulations are conducted, where damage is modeled by abrupt stiffness variation in the time-varying system. Experimental and real-world seismic-excited structure examples with time-varying stiffness are also presented to illustrate the capability of the developed WT–ICA method. Results show that the WT–ICA algorithm realizes accurate and robust blind identification of damage instant and location in single or multiple damage events. [Copyright &y& Elsevier]

• The transition from traditional methods to ML and DL has never been discussed for vibration-based SDD. • This paper aims to fulfill this gap. • Traditional methods and comprehensive review of the modern ML and DL use are presented. • The review is focusing on the vibration-based structural damage detection in civil structures only. Monitoring structural damage is extremely important for sustaining and preserving the service life of civil structures. While successful monitoring provides resolute and staunch information on the health, serviceability, integrity and safety of structures; maintaining continuous performance of a structure depends highly on monitoring the occurrence, formation and propagation of damage. Damage may accumulate on structures due to different environmental and human-induced factors. Numerous monitoring and detection approaches have been developed to provide practical means for early warning against structural damage or any type of anomaly. Considerable effort has been put into vibration-based methods, which utilize the vibration response of the monitored structure to assess its condition and identify structural damage. Meanwhile, with emerging computing power and sensing technology in the last decade, Machine Learning (ML) and especially Deep Learning (DL) algorithms have become more feasible and extensively used in vibration-based structural damage detection with elegant performance and often with rigorous accuracy. While there have been multiple review studies published on vibration-based structural damage detection, there has not been a study where the transition from traditional methods to ML and DL methods are described and discussed. This paper aims to fulfill this gap by presenting the highlights of the traditional methods and provide a comprehensive review of the most recent applications of ML and DL algorithms utilized for vibration-based structural damage detection in civil structures. [ABSTRACT FROM AUTHOR]

• Statistical pattern recognition identifies patterns to assess the presence of damage. • Features in quefrency domain are analytically derived from structural properties. • PCA projection increases features robustness to external distrubances. • New features are computationally advantageous compared to natural frequencies. • Proposed damage assessment strategy is tested with numerical and experimental data. The difficulties in dynamic system modeling and the recently developed machine learning tools have moved the attention of the structural damage assessment community towards a new direction. Attention is now placed on the immediate extraction of damage sensitive features directly from the monitored dynamic response of the structure, allowing engineers by-pass the creation of complex and sophisticated structural models. It is in this framework that the work presented in this paper has to be considered: the extraction of new damage sensitive features in the quefrency domain is discussed in the context of a vibration based analysis. Despite such a domain has been fully explored in acoustics, it represents an uncommon approach for the civil engineering community. In this paper, some variables characterizing the system dynamic response in the quefrency domain, referred to as cepstral coefficients, have been derived as functions of the structural characteristics and used in a damage assessment strategy. The main advantage of the quefrency domain over the frequency domain consists in a massive dimensionality reduction and, when dealing with real data, in a simplification of the extraction of cepstral coefficients versus that of natural frequencies. In this paper, the analytical expression of cepstral coefficients of the structural acceleration response has been derived and the statistical distribution of such coefficients has been investigated in a pattern recognition approach. In addition, a damage assessment strategy has been proposed by extracting, through a Principal Component Analysis (PCA), the projections of the cepstral coefficients with the lowest variances so to minimize the effect of external disturbances. The effectiveness of the proposed damage assessment strategy has been validated through numerical simulations and experimental data. [ABSTRACT FROM AUTHOR]